Calculate speed rotation with OMRON PLC
- Thread starter smarius
- Start date
kamenges
Member
I'm happy for you. So....what's your question?
Keith
Keith
How i can do this?
i try PRV2 but the speed oscillating
lostcontrol
Lifetime Supporting Member
By how much is it oscillating? I doubt that you will get a steady speed reading (such as the same reading scan after scan).
How do you know that the load speed is steady?
How do you know that the load speed is steady?
rotation speed oscillate beetwen 490 and 530 rpm
i use a optical sensor who read 18 pulse per revolution
i know it steady because i measure the rotation speed with a hand tahometer and frequency is stable at 50hz (500 Rpm)
Sorry for my english
thank you
Paul B
Member
I'm guessing you miss one pulse and this causes the erratic display between 490 and 530 rpm.
If you want to monitor RPM and want a stable reading you must do it over a long period of time. In other words you can't miss the pulse.
If you want to monitor RPM and stop a process when it slows down you must do it over a short period of time. You have to balance your sample rate so you catch the condition fast enough without over-correcting. Then add a rung of logic that says it has to be low for x.xx seconds so you will stop when its slow for more than the one scan, more than the one pulse you are missing.
In my experience it is useful to have two separate routines. One is fast enough to catch the slowdown and stop the process. The other looks pretty for the display. Then you can control and it looks pretty.
If you want to monitor RPM and want a stable reading you must do it over a long period of time. In other words you can't miss the pulse.
If you want to monitor RPM and stop a process when it slows down you must do it over a short period of time. You have to balance your sample rate so you catch the condition fast enough without over-correcting. Then add a rung of logic that says it has to be low for x.xx seconds so you will stop when its slow for more than the one scan, more than the one pulse you are missing.
In my experience it is useful to have two separate routines. One is fast enough to catch the slowdown and stop the process. The other looks pretty for the display. Then you can control and it looks pretty.
ok i understand, but the problem is that i want to be stable even when rpm grows up (at 80% of nominal RPM must connect a process)
Paul B
Member
Capturing a slowdown condition was just an example.
If you are counting pulses over a period of time, then calculating RPM, there will be an error. Your PLC logic evaluates pulses over time to calculate RPM periodically. So when it is time to calculate you might have 400 pulses this time, then 401 pulses the next time. So each time RPM is calculated it will "oscillate". The actual RPM is not oscillating - its the error.
You can minimize the error (minimize the oscillation) by calculating RPM less frequently so there are more pulses to evaluate. Beyond that, I don't know what to tell you other than what I already told you: Use a timer to filter action based on RPM, so the oscillations are ignored. Calculate rpm over a long period of time for display purposes so it looks pretty.
If you are counting pulses over a period of time, then calculating RPM, there will be an error. Your PLC logic evaluates pulses over time to calculate RPM periodically. So when it is time to calculate you might have 400 pulses this time, then 401 pulses the next time. So each time RPM is calculated it will "oscillate". The actual RPM is not oscillating - its the error.
You can minimize the error (minimize the oscillation) by calculating RPM less frequently so there are more pulses to evaluate. Beyond that, I don't know what to tell you other than what I already told you: Use a timer to filter action based on RPM, so the oscillations are ignored. Calculate rpm over a long period of time for display purposes so it looks pretty.
OkiePC
Lifetime Supporting Member
In a situation like this, I normally will measure the time between pulses (seconds per pulse) and then compute a near real time value in the desired engineering units.
Then I will average several of them together to get a more stable result. I will adjust how many samples are averaged to make the best compromise between response time and stability.
Sometimes, you can get away with averaging only 3 or four samples, other times, you may need a dozen or more samples to get the result you want. I know nothing about Omron PLCs so I can't tell you specifics about how to apply this, only the general concept.
Paul
Then I will average several of them together to get a more stable result. I will adjust how many samples are averaged to make the best compromise between response time and stability.
Sometimes, you can get away with averaging only 3 or four samples, other times, you may need a dozen or more samples to get the result you want. I know nothing about Omron PLCs so I can't tell you specifics about how to apply this, only the general concept.
Paul
Peter Nachtwey
Member
Why don't you guys do the math?
500 rpm = 8.333 rps
8.333 rps x 18 ppr=150 pps
or 1 pulse every 6.6 ms.
Now do you really think you can measure the time between pulses without a high speed counter card?
At 500 rpm I would count the pulse per second.
500 rpm = 8.333 rps
8.333 rps x 18 ppr=150 pps
or 1 pulse every 6.6 ms.
Now do you really think you can measure the time between pulses without a high speed counter card?
At 500 rpm I would count the pulse per second.
youse interubt
talking mpu. programing you youse a' interubt produced from the rotery incoder, ther ise yousely a. 1 bt signal each rotation, then start cont in a knoven tim an then doo the mathe like Peter are talking about....
the omron cqm1 unit have 2 interubt adr. om 000 and 003 i vould trye that, not schuer hou it works..
best kim jessen
sory mey englisch...
talking mpu. programing you youse a' interubt produced from the rotery incoder, ther ise yousely a. 1 bt signal each rotation, then start cont in a knoven tim an then doo the mathe like Peter are talking about....
the omron cqm1 unit have 2 interubt adr. om 000 and 003 i vould trye that, not schuer hou it works..
best kim jessen
sory mey englisch...
BITS N BYTES
Lifetime Supporting Member
Which model PLC?
Over what time base do you want the results? Every second/every 10 seconds/once per minute......
Over what time base do you want the results? Every second/every 10 seconds/once per minute......
Similar Topics
Hello,
I'm trying to understand a servo motion system. I want to know the motor speed in RPM at a given speed setpoint entered in a MAJ...
We like to know the speed of a vehicle, it‘s a normal car. Two photoelectric sensors (mounted only 1 meter distance from each other) are connected...
Hello
First this ia really great site. Glad i found it.
Here is my question.
I need to calculate the full speed of a driven auger that is...
Gear Reduction Box = 3.0
Drive Sprocket = 22 Teeth
Driven Sprocket = 88 Teeth
VFD = PowerFlex 40
I have a VFD controlling the speed of a...
Hi,
I am looking for a little help. I currently have an old Telemecanique TSx 17 that I need to replace. I am looking at installing an AB...
